@@ -1,6 +1,6 @@

 Package: ISAnalytics

 Title: Analyze gene therapy vector insertion sites data identified from genomics next generation sequencing reads for clonal tracking studies

-Version: 1.3.6

+Version: 1.3.7

 Date: 2020-07-03

 Authors@R: c(

   person(given = "Andrea",

@@ -34,6 +34,7 @@ export(import_parallel_Vispa2Matrices_interactive)

 export(import_single_Vispa2Matrix)

 export(integration_alluvial_plot)

 export(is_sharing)

+export(iss_source)

 export(known_clinical_oncogenes)

 export(mandatory_IS_vars)

 export(matching_options)

@@ -2,6 +2,16 @@

 title: "NEWS"

 output: github_document

 ---

+# ISAnalytics 1.3.7 (2021-10-20)

+

+## NEW

+

+* Added new feature `iss_source()`

+

+## FIXES

+

+* Fixed minor issues in data files `refGenes_mm9` and function `compute_near_integrations()`

+

 # ISAnalytics 1.3.6 (2021-10-05)

 ## NEW

@@ -1,6 +1,17 @@

 NEWS

 ================

+# ISAnalytics 1.3.7 (2021-10-20)

+

+## NEW

+

+-   Added new feature `iss_source()`

+

+## FIXES

+

+-   Fixed minor issues in data files `refGenes_mm9` and function

+    `compute_near_integrations()`

+

 # ISAnalytics 1.3.6 (2021-10-05)

 ## NEW

@@ -1545,21 +1545,25 @@ is_sharing <- function(...,

 #'     association_file = rlang::current_env()$association_file,

 #'     value_cols = c("seqCount", "fragmentEstimate")

 #' )

-#' filtered_by_purity <- purity_filter(x = aggreg,

-#' value_column = "seqCount_sum")

+#' filtered_by_purity <- purity_filter(

+#'     x = aggreg,

+#'     value_column = "seqCount_sum"

+#' )

 #' head(filtered_by_purity)

 purity_filter <- function(x,

-                          lineages = blood_lineages_default(),

-                          aggregation_key = c("SubjectID", "CellMarker",

-                                              "Tissue", "TimePoint"),

-                          group_key = c("CellMarker", "Tissue"),

-                          selected_groups = NULL,

-                          join_on = "CellMarker",

-                          min_value = 3,

-                          impurity_threshold = 10,

-                          by_timepoint = TRUE,

-                          timepoint_column = "TimePoint",

-                          value_column = "seqCount_sum") {

+    lineages = blood_lineages_default(),

+    aggregation_key = c(

+        "SubjectID", "CellMarker",

+        "Tissue", "TimePoint"

+    ),

+    group_key = c("CellMarker", "Tissue"),

+    selected_groups = NULL,

+    join_on = "CellMarker",

+    min_value = 3,

+    impurity_threshold = 10,

+    by_timepoint = TRUE,

+    timepoint_column = "TimePoint",

+    value_column = "seqCount_sum") {

     ## Checks

     #### - Base

     stopifnot(is.data.frame(x))

@@ -1570,12 +1574,12 @@ purity_filter <- function(x,

     stopifnot(is.numeric(impurity_threshold) || is.integer(impurity_threshold))

     stopifnot(is.character(value_column))

     stopifnot(is.null(selected_groups) || is.character(selected_groups) ||

-                  is.data.frame(selected_groups))

+        is.data.frame(selected_groups))

     #### - Keys

     if (!all(aggregation_key %in% colnames(x))) {

         rlang::abort(.missing_user_cols_error(

             aggregation_key[!aggregation_key %in% colnames(x)]

-            ))

+        ))

     }

     if (!value_column[1] %in% colnames(x)) {

         rlang::abort(.missing_user_cols_error(value_column))

@@ -1588,23 +1592,28 @@ purity_filter <- function(x,

         ### If lineages info is needed

         stopifnot(is.data.frame(lineages))

         if (!all(group_key %in% unique(c(colnames(x), colnames(lineages))))) {

-            missing_cols <- group_key[!group_key %in% unique(c(colnames(x),

-                                               colnames(lineages)))]

+            missing_cols <- group_key[!group_key %in% unique(c(

+                colnames(x),

+                colnames(lineages)

+            ))]

             rlang::abort(.missing_user_cols_error(missing_cols))

         }

         if (!(all(join_on %in% colnames(x)) &

-              all(join_on %in% colnames(lineages)))

-            ) {

+            all(join_on %in% colnames(lineages)))

+        ) {

             missing_common <- c("Missing common column(s) to join on",

-                                i = paste("The column(s) provided in argument",

-                                          "`join_on` is missing from one or",

-                                          "both data frames, aborting"))

-           rlang::abort(missing_common)

+                i = paste(

+                    "The column(s) provided in argument",

+                    "`join_on` is missing from one or",

+                    "both data frames, aborting"

+                )

+            )

+            rlang::abort(missing_common)

         }

         TRUE

     }

     if (by_timepoint) {

-       stopifnot(is.character(timepoint_column))

+        stopifnot(is.character(timepoint_column))

         timepoint_column <- timepoint_column[1]

         if (!timepoint_column %in% colnames(x)) {

             rlang::abort(.missing_user_cols_error(timepoint_column))

@@ -1625,13 +1634,17 @@ purity_filter <- function(x,

     }

     grouped <- x %>%

         dplyr::group_by(dplyr::across(dplyr::all_of(c(is_vars, group_key)))) %>%

-        dplyr::summarise(Value = sum(.data[[value_column]]),

-                         .groups = "drop")

+        dplyr::summarise(

+            Value = sum(.data[[value_column]]),

+            .groups = "drop"

+        )

     #### - value filter

-    filtered_value <- threshold_filter(x = grouped,

-                                 threshold = min_value,

-                                 cols_to_compare = "Value",

-                                 comparators = ">=")

+    filtered_value <- threshold_filter(

+        x = grouped,

+        threshold = min_value,

+        cols_to_compare = "Value",

+        comparators = ">="

+    )

     #### - Separating IS 1: group filtering

     pre_filt <- list()

     if (is.null(selected_groups) || purrr::is_empty(selected_groups)) {

@@ -1644,7 +1657,7 @@ purity_filter <- function(x,

             dplyr::filter(!.data[[group_key[1]]] %in% selected_groups)

     } else {

         ok_cols <- colnames(selected_groups)[colnames(selected_groups) %in%

-                                                 group_key]

+            group_key]

         selected_groups <- selected_groups %>%

             dplyr::select(dplyr::all_of(ok_cols)) %>%

             dplyr::distinct()

@@ -1654,11 +1667,13 @@ purity_filter <- function(x,

             pre_filt[["keep"]] <- filtered_value[0, ]

         } else {

             pre_filt[["process"]] <- dplyr::inner_join(filtered_value,

-                                                       selected_groups,

-                                                       by = ok_cols)

+                selected_groups,

+                by = ok_cols

+            )

             pre_filt[["keep"]] <- dplyr::anti_join(filtered_value,

-                                                   selected_groups,

-                                                   by = ok_cols)

+                selected_groups,

+                by = ok_cols

+            )

         }

     }

     if (nrow(pre_filt$process) == 0) {

@@ -1695,7 +1710,7 @@ purity_filter <- function(x,

         max_val <- max(group$Value)

         processed <- group %>%

             dplyr::mutate(remove = (max_val / .data$Value) >

-                              impurity_threshold) %>%

+                impurity_threshold) %>%

             dplyr::filter(remove == FALSE) %>%

             dplyr::select(-.data$remove)

         processed

@@ -1711,6 +1726,163 @@ purity_filter <- function(x,

     final

 }

+#' Find the source of IS by evaluating sharing.

+#'

+#' @description \lifecycle{experimental}

+#' The function computes the sharing between a reference group of interest

+#' for each time point and a selection of groups of interest. In this way

+#' it is possible to observe the percentage of shared integration sites between

+#' reference and each group and identify in which time point a certain IS was

+#' observed for the first time.

+#'

+#' @param reference A data frame containing one or more groups of reference.

+#' Groups are identified by `ref_group_key`

+#' @param selection A data frame containing one or more groups of interest

+#' to compare.

+#' Groups are identified by `selection_group_key`

+#' @param ref_group_key Character vector of column names that identify a

+#' unique group in the `reference` data frame

+#' @param selection_group_key Character vector of column names that identify a

+#' unique group in the `selection` data frame

+#' @param timepoint_column Name of the column holding time point

+#' info?

+#' @param by_subject Should calculations be performed for each subject

+#' separately?

+#' @param subject_column Name of the column holding subjects information.

+#' Relevant only if `by_subject = TRUE`

+#'

+#' @return A list of data frames or a data frame

+#' @family Analysis functions

+#' @export

+#'

+#' @examples

+#' data("integration_matrices", package = "ISAnalytics")

+#' data("association_file", package = "ISAnalytics")

+#' aggreg <- aggregate_values_by_key(

+#'     x = rlang::current_env()$integration_matrices,

+#'     association_file = rlang::current_env()$association_file,

+#'     value_cols = c("seqCount", "fragmentEstimate")

+#' )

+#' df1 <- aggreg %>%

+#'     dplyr::filter(.data$Tissue == "BM")

+#' df2 <- aggreg %>%

+#'     dplyr::filter(.data$Tissue == "PB")

+#' source <- iss_source(df1, df2)

+#' source

+#' ggplot2::ggplot(source$PT001, ggplot2::aes(

+#'     x = as.factor(g2_TimePoint),

+#'     y = sharing_perc, fill = g1

+#' )) +

+#'     ggplot2::geom_col() +

+#'     ggplot2::labs(

+#'         x = "Time point", y = "Shared IS % with MNC BM",

+#'         title = "Source of is MNC BM vs MNC PB"

+#'     )

+iss_source <- function(reference,

+    selection,

+    ref_group_key = c(

+        "SubjectID", "CellMarker",

+        "Tissue", "TimePoint"

+    ),

+    selection_group_key = c(

+        "SubjectID", "CellMarker",

+        "Tissue", "TimePoint"

+    ),

+    timepoint_column = "TimePoint",

+    by_subject = TRUE,

+    subject_column = "SubjectID") {

+    ## Checks

+    stopifnot(is.data.frame(reference) & is.data.frame(selection))

+    stopifnot(is.character(ref_group_key) & is.character(selection_group_key))

+    stopifnot(is.character(timepoint_column))

+    stopifnot(is.logical(by_subject))

+    by_subject <- by_subject[1]

+    if (!all(ref_group_key %in% colnames(reference))) {

+        rlang::abort(.missing_user_cols_error(

+            ref_group_key[!ref_group_key %in% colnames(reference)]

+        ))

+    }

+    if (!all(selection_group_key %in% colnames(selection))) {

+        rlang::abort(.missing_user_cols_error(

+            selection_group_key[!selection_group_key %in% colnames(selection)]

+        ))

+    }

+    timepoint_column <- timepoint_column[1]

+    if (!timepoint_column %in% colnames(reference) |

+        !timepoint_column %in% colnames(selection)) {

+        rlang::abort(.missing_needed_cols(timepoint_column))

+    }

+    ## Workflow choice

+    if (by_subject) {

+        stopifnot(is.character(subject_column))

+        subject_column <- subject_column[1]

+        ref_split <- reference %>%

+            dplyr::group_by(dplyr::across({{ subject_column }}))

+        ref_subjs <- ref_split %>%

+            dplyr::group_keys() %>%

+            dplyr::pull(.data[[subject_column]])

+        ref_split <- ref_split %>%

+            dplyr::group_split() %>%

+            purrr::set_names(ref_subjs)

+        sel_split <- selection %>%

+            dplyr::group_by(dplyr::across({{ subject_column }}))

+        sel_subjs <- sel_split %>%

+            dplyr::group_keys() %>%

+            dplyr::pull(.data[[subject_column]])

+        sel_split <- sel_split %>%

+            dplyr::group_split() %>%

+            purrr::set_names(sel_subjs)

+        shared <- .sharing_for_source(ref_split,

+            sel_split,

+            ref_key = ref_group_key,

+            sel_key = selection_group_key,

+            tp_col = timepoint_column,

+            subj_col = subject_column

+        )

+        shared <- purrr::map(

+            shared,

+            ~ .x %>%

+                dplyr::select(

+                    -.data$count_g1, -.data$count_g2,

+                    -.data$count_union

+                ) %>%

+                tidyr::unnest(.data$is_coord, keep_empty = TRUE) %>%

+                dplyr::mutate(sharing_perc = dplyr::if_else(

+                    shared == 0, 0, .data$on_g2 / .data$shared

+                )) %>%

+                dplyr::select(

+                    -.data$shared, -.data$on_g1, -.data$on_g2,

+                    -.data$on_union

+                )

+        )

+    } else {

+        shared <- .sharing_for_source(reference,

+            selection,

+            ref_key = ref_group_key,

+            sel_key = selection_group_key,

+            tp_col = timepoint_column,

+            subj_col = subject_column

+        )

+        shared <- shared %>%

+            dplyr::select(

+                -.data$count_g1, -.data$count_g2,

+                -.data$count_union

+            ) %>%

+            tidyr::unnest(.data$is_coord, keep_empty = TRUE) %>%

+            dplyr::mutate(sharing_perc = dplyr::if_else(shared == 0,

+                0,

+                .data$on_g2 /

+                    .data$shared

+            )) %>%

+            dplyr::select(

+                -.data$shared, -.data$on_g1, -.data$on_g2,

+                -.data$on_union

+            )

+    }

+

+    return(shared)

+}

+

 #' A set of pre-defined functions for `sample_statistics`.

 #'

 #' @return A named list of functions/purrr-style lambdas

@@ -4735,3 +4735,170 @@

     }

     sharing_df

 }

+

+#---- USED IN : iss_source ----

+.assign_iss_by_tp <- function(df, timepoint_column) {

+    is_vars <- if (.is_annotated(df)) {

+        c(mandatory_IS_vars(), annotation_IS_vars())

+    } else {

+        mandatory_IS_vars()

+    }

+    return(df %>%

+        dplyr::mutate(!!timepoint_column := as.numeric(

+            .data[[timepoint_column]]

+        )) %>%

+        dplyr::group_by(dplyr::across(dplyr::all_of(is_vars))) %>%

+        dplyr::group_modify(~ {

+            if (nrow(.x) == 1) {

+                return(.x)

+            }

+            min_tp <- min(.x[[timepoint_column]])

+            return(.x %>%

+                dplyr::filter(.data[[timepoint_column]] == min_tp))

+        }) %>%

+        dplyr::ungroup())

+}

+

+.sharing_for_source <- function(ref,

+    sel,

+    ref_key,

+    sel_key,

+    tp_col,

+    subj_col) {

+    if (!is.data.frame(ref)) {

+        ### Workflow by subject

+        common_names <- if (!all(names(ref) == names(sel))) {

+            intersect(names(ref), names(sel))

+        } else {

+            names(ref)

+        }

+        if (length(common_names) == 0) {

+            no_common_err <- c("No common subjects",

+                x = paste(

+                    "No common subjects between",

+                    "reference and selection"

+                ),

+                i = paste(

+                    "In reference:",

+                    paste0(names(ref), collapse = ", "),

+                    "\n",

+                    "In selection: ",

+                    paste0(names(sel), collapse = ", ")

+                )

+            )

+            rlang::abort(no_common_err)

+        }

+        if (getOption("ISAnalytics.verbose") == TRUE &&

+            (length(common_names) < length(names(ref)) ||

+                length(common_names) < length(names(sel)))) {

+            all_names <- union(names(ref), names(sel))

+            common_warn_msg <- c("Mismatch in subjects found",

+                i = paste(

+                    "Some subjects were excluded from",

+                    "computations because they were",

+                    "absent from reference or from",

+                    "selection"

+                ),

+                paste(

+                    "Excluded: ",

+                    paste0(all_names[!all_names %in%

+                        common_names],

+                    collapse = ", "

+                    )

+                )

+            )

+        }

+        if (.Platform$OS.type == "windows") {

+            p <- BiocParallel::SnowParam(

+                tasks = length(common_names),

+                progressbar = getOption("ISAnalytics.verbose"),

+                exportglobals = FALSE

+            )

+        } else {

+            p <- BiocParallel::MulticoreParam(

+                tasks = length(common_names),

+                progressbar = getOption("ISAnalytics.verbose"),

+                exportglobals = FALSE

+            )

+        }

+        FUN <- function(subj,

+    ref, sel, ref_key, sel_key,

+    tp_col) {

+            ref_df <- ref[[subj]]

+            sel_df <- sel[[subj]]

+            ref_key_min <- ref_key[ref_key != tp_col]

+            ref_split <- ref_df %>%

+                dplyr::group_by(dplyr::across(dplyr::all_of(ref_key_min))) %>%

+                dplyr::group_split()

+            quiet_sharing <- purrr::quietly(is_sharing)

+            sharing <- purrr::map_df(ref_split, ~ {

+                assigned_ref <- .assign_iss_by_tp(.x,

+                    timepoint_column = tp_col

+                )

+                sh <- (quiet_sharing(assigned_ref,

+                    sel_df,

+                    group_keys = list(

+                        g1 = ref_key,

+                        g2 = sel_key

+                    ),

+                    keep_genomic_coord = TRUE

+                ))$result

+                sh <- sh %>%

+                    tidyr::separate(

+                        col = "g1",

+                        into = paste0("g1_", ref_key),

+                        remove = FALSE,

+                        convert = TRUE

+                    ) %>%

+                    tidyr::separate(

+                        col = "g2",

+                        into = paste0("g2_", sel_key),

+                        remove = FALSE,

+                        convert = TRUE

+                    )

+            })

+        }

+

+        shared <- BiocParallel::bplapply(common_names, FUN,

+            ref = ref,

+            sel = sel,

+            ref_key = ref_key,

+            sel_key = sel_key,

+            tp_col = tp_col,

+            BPPARAM = p

+        ) %>%

+            purrr::set_names(common_names)

+        return(shared)

+    }

+    ref_key_min <- ref_key[ref_key != tp_col]

+    ref_split <- ref %>%

+        dplyr::group_by(dplyr::across(dplyr::all_of(ref_key_min))) %>%

+        dplyr::group_split()

+    quiet_sharing <- purrr::quietly(is_sharing)

+    sharing <- purrr::map_df(ref_split, ~ {

+        assigned_ref <- .assign_iss_by_tp(.x,

+            timepoint_column = tp_col

+        )

+        (quiet_sharing(assigned_ref,

+            sel,

+            group_keys = list(

+                g1 = ref_key,

+                g2 = sel_key

+            ),

+            keep_genomic_coord = TRUE

+        ))$result

+    })

+    sharing %>%

+        tidyr::separate(

+            col = "g1",

+            into = paste0("g1_", ref_key),

+            remove = FALSE,

+            convert = TRUE

+        ) %>%

+        tidyr::separate(

+            col = "g2",

+            into = paste0("g2_", sel_key),

+            remove = FALSE,

+            convert = TRUE

+        )

+}

@@ -137,6 +137,17 @@ options("ISAnalytics.reports" = TRUE)

 Show more

 </summary>

+# ISAnalytics 1.3.7 (2021-10-20)

+

+## NEW

+

+-   Added new feature `iss_source()`

+

+## FIXES

+

+-   Fixed minor issues in data files `refGenes_mm9` and function

+    `compute_near_integrations()`

+

 # ISAnalytics 1.3.6 (2021-10-05)

 ## NEW

@@ -82,6 +82,7 @@ reference:

   - cumulative_is

   - is_sharing

   - purity_filter

+  - iss_source

 - title: "HSC population estimate"

 - contents:

   - HSC_population_size_estimate

@@ -80,6 +80,7 @@ Other Analysis functions:

 \code{\link{cumulative_count_union}()},

 \code{\link{cumulative_is}()},

 \code{\link{is_sharing}()},

+\code{\link{iss_source}()},

 \code{\link{purity_filter}()},

 \code{\link{sample_statistics}()},

 \code{\link{separate_quant_matrices}()},

@@ -70,6 +70,7 @@ Other Analysis functions:

 \code{\link{cumulative_count_union}()},

 \code{\link{cumulative_is}()},

 \code{\link{is_sharing}()},

+\code{\link{iss_source}()},

 \code{\link{purity_filter}()},

 \code{\link{sample_statistics}()},

 \code{\link{separate_quant_matrices}()},

@@ -62,6 +62,7 @@ Other Analysis functions:

 \code{\link{cumulative_count_union}()},

 \code{\link{cumulative_is}()},

 \code{\link{is_sharing}()},

+\code{\link{iss_source}()},

 \code{\link{purity_filter}()},

 \code{\link{sample_statistics}()},

 \code{\link{separate_quant_matrices}()},

@@ -91,6 +91,7 @@ Other Analysis functions:

 \code{\link{compute_abundance}()},

 \code{\link{cumulative_is}()},

 \code{\link{is_sharing}()},

+\code{\link{iss_source}()},

 \code{\link{purity_filter}()},

 \code{\link{sample_statistics}()},

 \code{\link{separate_quant_matrices}()},

@@ -57,6 +57,7 @@ Other Analysis functions:

 \code{\link{compute_abundance}()},

 \code{\link{cumulative_count_union}()},

 \code{\link{is_sharing}()},

+\code{\link{iss_source}()},

 \code{\link{purity_filter}()},

 \code{\link{sample_statistics}()},

 \code{\link{separate_quant_matrices}()},

@@ -95,6 +95,7 @@ Other Analysis functions:

 \code{\link{compute_abundance}()},

 \code{\link{cumulative_count_union}()},

 \code{\link{cumulative_is}()},

+\code{\link{iss_source}()},

 \code{\link{purity_filter}()},

 \code{\link{sample_statistics}()},

 \code{\link{separate_quant_matrices}()},

new file mode 100644

@@ -0,0 +1,85 @@

+% Generated by roxygen2: do not edit by hand

+% Please edit documentation in R/analysis-functions.R

+\name{iss_source}

+\alias{iss_source}

+\title{Find the source of IS by evaluating sharing.}

+\usage{

+iss_source(

+  reference,

+  selection,

+  ref_group_key = c("SubjectID", "CellMarker", "Tissue", "TimePoint"),

+  selection_group_key = c("SubjectID", "CellMarker", "Tissue", "TimePoint"),

+  timepoint_column = "TimePoint",

+  by_subject = TRUE,

+  subject_column = "SubjectID"

+)

+}

+\arguments{

+\item{reference}{A data frame containing one or more groups of reference.

+Groups are identified by \code{ref_group_key}}

+

+\item{selection}{A data frame containing one or more groups of interest

+to compare.

+Groups are identified by \code{selection_group_key}}

+

+\item{ref_group_key}{Character vector of column names that identify a

+unique group in the \code{reference} data frame}

+

+\item{selection_group_key}{Character vector of column names that identify a

+unique group in the \code{selection} data frame}

+

+\item{timepoint_column}{Name of the column holding time point

+info?}

+

+\item{by_subject}{Should calculations be performed for each subject

+separately?}

+

+\item{subject_column}{Name of the column holding subjects information.

+Relevant only if \code{by_subject = TRUE}}

+}

+\value{

+A list of data frames or a data frame

+}

+\description{

+\lifecycle{experimental}

+The function computes the sharing between a reference group of interest

+for each time point and a selection of groups of interest. In this way

+it is possible to observe the percentage of shared integration sites between

+reference and each group and identify in which time point a certain IS was

+observed for the first time.

+}

+\examples{

+data("integration_matrices", package = "ISAnalytics")

+data("association_file", package = "ISAnalytics")

+aggreg <- aggregate_values_by_key(

+    x = rlang::current_env()$integration_matrices,

+    association_file = rlang::current_env()$association_file,

+    value_cols = c("seqCount", "fragmentEstimate")

+)

+df1 <- aggreg \%>\%

+    dplyr::filter(.data$Tissue == "BM")

+df2 <- aggreg \%>\%

+    dplyr::filter(.data$Tissue == "PB")

+source <- iss_source(df1, df2)

+source

+ggplot2::ggplot(source$PT001, ggplot2::aes(x = as.factor(g2_TimePoint),

+                                           y = sharing_perc, fill = g1)) +

+    ggplot2::geom_col() +

+    ggplot2::labs(x = "Time point", y = "Shared IS \% with MNC BM",

+                  title = "Source of is MNC BM vs MNC PB")

+}

+\seealso{

+Other Analysis functions: 

+\code{\link{CIS_grubbs}()},

+\code{\link{comparison_matrix}()},

+\code{\link{compute_abundance}()},

+\code{\link{cumulative_count_union}()},

+\code{\link{cumulative_is}()},

+\code{\link{is_sharing}()},

+\code{\link{purity_filter}()},

+\code{\link{sample_statistics}()},

+\code{\link{separate_quant_matrices}()},

+\code{\link{threshold_filter}()},

+\code{\link{top_integrations}()}

+}

+\concept{Analysis functions}

@@ -108,8 +108,10 @@ aggreg <- aggregate_values_by_key(

     association_file = rlang::current_env()$association_file,

     value_cols = c("seqCount", "fragmentEstimate")

 )

-filtered_by_purity <- purity_filter(x = aggreg,

-value_column = "seqCount_sum")

+filtered_by_purity <- purity_filter(

+    x = aggreg,

+    value_column = "seqCount_sum"

+)

 head(filtered_by_purity)

 }

 \seealso{

@@ -120,6 +122,7 @@ Other Analysis functions:

 \code{\link{cumulative_count_union}()},

 \code{\link{cumulative_is}()},

 \code{\link{is_sharing}()},

+\code{\link{iss_source}()},

 \code{\link{sample_statistics}()},

 \code{\link{separate_quant_matrices}()},

 \code{\link{threshold_filter}()},

@@ -78,6 +78,7 @@ Other Analysis functions:

 \code{\link{cumulative_count_union}()},

 \code{\link{cumulative_is}()},

 \code{\link{is_sharing}()},

+\code{\link{iss_source}()},

 \code{\link{purity_filter}()},

 \code{\link{separate_quant_matrices}()},

 \code{\link{threshold_filter}()},

@@ -62,6 +62,7 @@ Other Analysis functions:

 \code{\link{cumulative_count_union}()},

 \code{\link{cumulative_is}()},

 \code{\link{is_sharing}()},

+\code{\link{iss_source}()},

 \code{\link{purity_filter}()},

 \code{\link{sample_statistics}()},

 \code{\link{threshold_filter}()},

@@ -211,6 +211,7 @@ Other Analysis functions:

 \code{\link{cumulative_count_union}()},

 \code{\link{cumulative_is}()},

 \code{\link{is_sharing}()},

+\code{\link{iss_source}()},

 \code{\link{purity_filter}()},

 \code{\link{sample_statistics}()},

 \code{\link{separate_quant_matrices}()},

@@ -78,6 +78,7 @@ Other Analysis functions:

 \code{\link{cumulative_count_union}()},

 \code{\link{cumulative_is}()},

 \code{\link{is_sharing}()},

+\code{\link{iss_source}()},

 \code{\link{purity_filter}()},

 \code{\link{sample_statistics}()},

 \code{\link{separate_quant_matrices}()},

@@ -947,3 +947,572 @@ test_that("cumulative_is produces correct output", {

     expect_true(all(mandatory_IS_vars() %in% colnames(c_is)))

     expect_true(nrow(c_is) == sum(counts))

 })

+

+#------------------------------------------------------------------------------#

+# Test .assign_iss_by_tp

+#------------------------------------------------------------------------------#

+test_that(".assign_iss_by_tp assigns iss correctly", {

+    test_df <- tibble::tribble(

+        ~chr, ~integration_locus, ~strand, ~CellMarker, ~Tissue, ~TimePoint,

+        "1", 12345, "+", "CD34", "BM", "03",

+        "1", 12345, "+", "CD34", "BM", "01",

+        "1", 12345, "+", "CD34", "BM", "06",

+        "2", 54321, "-", "CD34", "BM", "01",

+        "3", 62135, "+", "CD34", "BM", "02",

+        "3", 62135, "+", "CD34", "BM", "08"

+    )

+

+    expected_assign <- tibble::tribble(

+        ~chr, ~integration_locus, ~strand, ~CellMarker, ~Tissue, ~TimePoint,

+        "1", 12345, "+", "CD34", "BM", 1,

+        "2", 54321, "-", "CD34", "BM", 1,

+        "3", 62135, "+", "CD34", "BM", 2

+    )

+    re_assigned <- .assign_iss_by_tp(

+        df = test_df,

+        timepoint_column = "TimePoint"

+    )

+    expect_equal(re_assigned, expected_assign)

+})

+

+#------------------------------------------------------------------------------#

+# Test iss_source

+#------------------------------------------------------------------------------#

+test_df2 <- tibble::tibble(

+    chr = c(

+        "1", "1", "1", "1", "2", "2", "3", "3",

+        "1", "1", "1", "1", "1", "2", "3", "3"

+    ),

+    integration_locus = c(

+        12345, 43524, 12345, 12345, 54321,

+        76835, 62135, 62135, 12345, 12345,

+        56832, 12345, 12345, 54321, 62135,

+        62135

+    ),

+    strand = c(

+        "+", "+", "+", "+", "-", "-", "+", "+", "+",

+        "+", "-", "+", "+", "-", "+", "+"

+    ),

+    SubjectID = c(

+        "PT01", "PT02", "PT01", "PT01", "PT01",

+        "PT02", "PT01", "PT01", "PT01", "PT01",

+        "PT02", "PT01", "PT02", "PT01", "PT01",

+        "PT01"

+    ),

+    CellMarker = c(

+        "CD34", "CD34", "CD34", "CD34", "CD34",

+        "CD34", "CD34", "CD34", "Whole", "Whole",

+        "Whole", "Whole", "Whole", "Whole", "Whole",

+        "Whole"

+    ),

+    Tissue = c(

+        "BM", "BM", "BM", "BM", "BM", "BM", "BM", "BM",

+        "BM", "BM", "BM", "BM", "BM", "BM", "BM", "BM"

+    ),

+    TimePoint = c(

+        "03", "03", "01", "06", "01", "01", "02",

+        "08", "03", "01", "01", "06", "06", "01",

+        "02", "08"

+    )

+)

+

+test_df3 <- tibble::tibble(

+    chr = c(

+        "1", "1", "1", "1", "2", "2", "3", "3", "1",

+        "1", "1", "1", "1", "2", "3", "3"

+    ),

+    integration_locus = c(

+        56252, 43524, 12345, 86435, 54321,

+        76835, 62135, 432245, 12345, 534587,

+        56832, 12345, 12345, 578635, 62135,

+        62135

+    ),

+    strand = c(

+        "+", "+", "+", "+", "-", "-", "+", "+", "+",

+        "+", "-", "+", "+", "-", "+", "+"

+    ),

+    SubjectID = c(

+        "PT02", "PT02", "PT01", "PT01", "PT01",

+        "PT02", "PT01", "PT01", "PT01", "PT01",

+        "PT02", "PT01", "PT02", "PT01", "PT02",

+        "PT01"

+    ),

+    CellMarker = c(

+        "CD14", "CD13", "CD15", "CD14", "CD14",

+        "CD13", "CD14", "CD13", "CD14", "CD15",

+        "CD15", "CD14", "CD15", "CD13", "CD14",

+        "CD13"

+    ),

+    Tissue = c(

+        "PB", "PB", "PB", "PB", "PB", "PB", "PB", "PB",

+        "PB", "PB", "PB", "PB", "PB", "PB", "PB", "PB"

+    ),

+    TimePoint = c(

+        "03", "03", "01", "06", "01", "01", "02",

+        "08", "03", "01", "01", "06", "06", "01",

+        "02", "08"